.
+ext = strrchr(buf, '.');
+ext = ext ? ext : strchr(buf, 0);
Use tcc_fileextension().
--- grischka
Kirill Smelkov wrote:
This affectes where `tcc -E -MD file.c` will place generated dependency
information -- previously, for `tcc -E` output_default was
yuanbin wrote:
if you don't like my patch, please recover
I fixed it (and two others too):
http://repo.or.cz/w/tinycc.git?a=shortlog;h=632ee5a5
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/ma
e "git gui" to enter commit messages easily.
Also you can use "gitk --all" to check your commits visually
before you push.
Also it is possible to use the "amend" option in "git gui" to
fix the last commit or to add some more changes.
Thanks,
--- grischk
us other people.
Thanks,
--- grischka
yuanbin wrote:
Coo - C, Object Oriented
http://sourceforge.net/projects/coo/
-coo.h--
#ifndef __COO_H__
#define __COO_H__
typedef struct VTable /*root of virtual table class*/
{
long offset; /*servers for FREE*/
} VTable;
#define EXTENDS
ored? how can i access them?
The parameters of the function macro symbol "sym" are linked in
"sym->next". See tccpp.c:2700:
sa = s->next;
and 2730:
sa = sa->next;
--- grischka
___
Tinycc-devel mailing list
Tinycc-deve
ld print as bogus anonymous label (at tccpp.c:338):
} else if (v >= SYM_FIRST_ANOM) {
/* special name for anonymous symbol */
sprintf(p, "L.%u", v - SYM_FIRST_ANOM);
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
eed random temporary filenames at all.
or at least it doesn't need to be as random as that. Something like tcc-
filename-$(date "+%s") would be enough I think, wouldn't it?
Usual (auto-)configure just use 'conftest.xxx' in current directory.
--- grischka
__
with MinGW/MSYS
which doesn't have /dev/urandom, dd, cksum and such.
Maybe we don't really need random temporary filenames at all.
--- grischka
Best regards,
Thomas Preud'homme
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.or
mobi phil wrote:
On Sun, May 16, 2010 at 10:02 PM, grischka wrote:
For bug reports please use latest mob or master branch from
http://repo.or.cz/w/tinycc.git
Thanks,
Did not mention it indeed, but this was tested on
git://repo.or.cz/tinycc.git, both master and mob.
Sorry. I think it
For bug reports please use latest mob or master branch from
http://repo.or.cz/w/tinycc.git
Thanks,
--- grischka
mobi phil wrote:
Hello,
the compiler throws invalid warnings when compiling pre-processed files:
I did
-> tcc -E main.c -o main.E.c
->#valid preprocessed main.
Someone please push this patch? Thanks,
--- grischka
Tomas Lindquist Olsen wrote:
Hi all.
I'm working on a compiler project where one of our targets is C, for
this I've found TinyCC very interesting and useful.
There was one issue though, which was not being able to (or at least
tuned to use a directory structure as it exists under win32.
But the win32 version of libtcc1.a should simply be win32/lib/libtcc1.a
and not some tcc1.def hack.
As to ARM, I think Daniel has or had a pre-built libtcc1.a for ARM
on his web-site (url can be found somewhere in this mail archive, IIRC)
may be parsed as part of a C identifier
2) p[-1] may refer to undefined memory
3) the feature (as in "jmp .L0") would not work anymore
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
Thomas Preud'homme wrote:
Hi all,
I'm sorry I made a little typo which made a huge error: it deleted the mob
branch.
Can happen.
I pushed my latest mob. I don't have a28b18fa though, so it's up to you
(or someone else) to push that and what's in between (if any
Detlef, please fix.
(Beware, not as suggested below ;))
--- grischka
Timo VJ Lähde wrote:
Maybe check first previous char if possible but using negative pointer
isn't safe:
tccpp.c line 2281
case '.':
/* check first for a local label (.Lxx:) */
if (p[1]
Please use the tinycc mailing list:
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
Original Message
Date: Wed, 28 Apr 2010 17:18:09 +0300
From: Kirill Joss
Hi !
Source code:
struct {
int DeviceID;
int StartAddr;
int Len;
int Data[32
-- Forwarded to the list --
From: Henry Kroll III
On Mon, 2010-04-26 at 01:08 +0200, grischka wrote:
Anyway, do what you need to feel comfortable, and don't forget to tell
us what new features we now have and how people can try them out, at
some point.
The new options
d how people can try them out, at
some point.
Thanks,
--- grischka
I think this last push fixed up
most of the major problems in my branch. This brings things in line with
the Fedora rpm I'm building.
http://thenerdshow.com/rpm/tinycc-git20100114-2.fc
ses them (from error(...)) and we
want tcc to be able to compile itself, as always.
Also, someone already gave us a configure switch "--with-libgcc"
(-> CONFIG_USE_LIBGCC), which might or might not be useful.
--- grischka
On the other hand, provide dummy functions is
perfectly possible
uot; this could easily mean several versions,
such as one for each gcc version that gnu/linux systems are compiled with.
Well, maybe I just don't understand. Can we make "tcc -static" work with
libgcc_eh.a, say somewhat easily?
--- grischka
Thomas Preud'homme
_
not our problem.
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
cc.git
fetch = +refs/heads/*:refs/remotes/origin/*
If you cloned from repo.or.cz initially, you already have that.
I use "gitk --all" and "git gui" for most things, and command line for
rebase or push.
--- grischka
--Timppa
Timo VJ Lähde wrote:
I prefer PE log file per file:
Sure.
(Of course we expect your work on the "mob" branch at some point,
but it's nice that you keep us informed what you're doing.)
--- grischka
--- a\tccpe.c Mon Apr 19 15:34:58 2010
+++ b\tccpe.c Wed Apr 21 09:1
reful not to mess up other peoples' work.
Don't worry, I know all the consequences it could produce.
Great ;)
--- grischka
Thomas Preud'homme
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
careful not to mess up other peoples' work.
Thomas Preud'homme
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
&libs, &nblibs, tcc_strdup(filename));
dynarray_add((void ***) &libs, &nblibs, tcc_strdup(libname));
But then again I'm a perfectionist ;)
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
it was not an error in GCC 3.x but is one in GCC 4.x. Maybe
we could make it just a warning for now. Do what you want ;)
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
x27;ll not write
the patch.
If I may propose something else then maybe you could just move the
default system path settings from tcc_new to tcc_set_output_type
(to where it happens for TCC_TARGET_PE) which is when the command
line was already parsed.
--- grischka
__
better
than nothing.
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
Forwarded to the mailing list.
--- grischka
Original Message
Subject: Re: [Tinycc-devel] Several patchs from Debian packaging
Date: Sun, 18 Apr 2010 06:40:47 +0200
From: RoboTux
Reply-To: robo...@celest.fr
To: grischka
References: <201004171452.32731.robo...@celest
. Maybe you can put the
library-names in a list (aka dynarray) and then just loop over that
list.
[P|e9406c0] Complain for static fct declared w/o file scope
Error out on static function without file scope and give an explaination to the
user
Probably too correct. It&
Timo VJ Lähde wrote:
Is that correct ?
tccpe.c 1824:
#if defined(TCC_TARGET_ARM)
pe.subsystem = 3;
#else
pe.subsystem = 9;
#endif
In WinCE ARM should be pe.subsystem = 9; ?
--Timppa
Looks wrong. Maybe it was meant to be
#if !defined(TCC_TARGET_ARM)
--- grischka
Kirill Smelkov wrote:
By the way, do you have an idea about when 0.9.26 is going to happen?
Even approximate timing helps...
Can happen whenever people want. You'd just agree on a revision
number that is reasonably stable and you can have it tomorrow ;)
--- gri
Timo VJ Lähde wrote:
If i compile tcc.exe with tcc.exe, i can't run it in WindowsXP because
of missing function in msvcrt.dll : ftime32().
--Timppa
Thanks. I've pushed a fix on "mob".
--- grischka
___
Tinycc-devel maili
I'm not sure whether Fabrice Bellard reads this but we can tell
him at next release.
Perhaps just
"Savannah project page and source repository"
would be more future safe ;)
--- grischka
Kirill Smelkov wrote:
It's been 1.5 years already [1], and Tinycc Savannah project
Vincent Lefevre wrote:
Another problem with tcc 0.9.25: the FLT_EVAL_METHOD macro (specified
by the ISO C standard) is not defined.
You can push patches on our "mob" branch:
http://repo.or.cz/w/tinycc.git
--- grischka
___
Ti
gest to rename "pe->type" to "pe->exetype" and then use
it to handle exe/dll, and use pe->subsystem to handle gui/console.
Note that "-WL,subsystem,xxx" should override the smart logic with
"WinMain". Also don't f
mpile win32/lib/wincrt1.c with
tcc.exe
Geez, just replace not only one but all the two
"ATTR_MODE(ad)"
by
"ad->mode"
.
Or discuss it with Detlef who wrote the bug.
--- grischka
___
Tinycc-devel mailing list
Timo VJ Lähde wrote:
No, but it makes possible to use _winstart as entrypoint directly
without runtime library.
You are repeating yourself.
--- grischka
- Original Message - From: "grischka"
To:
Sent: Friday, April 02, 2010 12:43 PM
Subject: Re: [Tinycc-devel] Wi
Please push this on "mob".
--- grischka
Timo VJ Lähde wrote:
win32/tools/tiny_libmaker leaves temporary files on error
-
--- a/tiny_libmaker.c Mon Mar 15 23:51:00 2010
+++ b/tiny_libmaker.c Wed Mar 31 22:35:06 201
:
case TOK_ATTRIBUTE2:
parse_attribute(ad);
-if (ATTR_MODE(ad)) {
+if (ad->mode) {
u = ATTR_MODE(ad) -1;
t = (t & ~VT_BTYPE) | u;
}
And what about the other occurrence of ATTR_MODE(ad) just below?
--- g
t;-Wl,--subsystem,gui" switch does not work, but how
does the patch below fix this?
--- grischka
--- test-gui.c ---
int _winstart(void)
{
return 0;
}
With this patch it's possible to use _winstart without runtime libraries:
tcc
don't have type.ref set. That's likely why it crashes
and wouldn't work anyway.
I know that the dllimport attribute works for data.
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
his more complex, that I've overlooked?
Maybe not. As always, please push patches on our "mob" branch ;)
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
Simon Lehmayr wrote:
Hi grischka,
no, I want to compile tcc from its sources and debug tcc under MS visual
studio.
Just as always: Create a console project, add tcc.c, fiddle with
the build options, and hit F7.
Well, you'd also need to create a config.h (see build-tcc.bat).
--- gri
Simon Lehmayr wrote:
Hi,
how can I setup tcc to compile and debug tcc itself with MS visual
studio 2005?
You want tcc do debug itself with MSVS?
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman
2/tcc-win32.txt:
Import Definition Files:
To link with Windows system DLLs, TCC uses import definition
files (.def) instead of libraries.
...
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
/archive/html/tinycc-devel/2010-02/
Btw, please don't send patches to the list. Instead push them
directly on our "mob" branch.
http://repo.or.cz/w/tinycc.git
Thanks,
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@
ONFIG_TCCDIR set correctly (and with i386-win32-tcc
installed) all you need (and ever should need) is:
$ cd win32/examples
$ i386-win32-tcc hello_win.c
This produces hello_win.exe that can be run on windows or under wine.
--- grischka
___
Tinycc-devel mai
CCDIR (config.h) individually, so you still
need to use the -B switch. Patches to our "mob" branch are welcome, as
always ;)
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
ch to
fill the GOT after linking (only for x86-64).
Ali
Please push patches on our "mob" branch:
http://repo.or.cz/w/tinycc.git
Thus it is also easier for other people to test and give you
feedback.
Thanks,
--- grischka
From: Ali Gholami Rudi
Subject: [PATCH] fill
. Is there any possibility,
a patch or is this a feature, which is planned?
What do you mean by "tcc obviously has no UNICODE support"?
I mean you can take hello_win.c, put #define UNICODE on top, make
some trivial changes and it works. What would you need more?
--
://msdn.microsoft.com/en-us/library/zthk2dkh.aspx
Do not expect me personally to fix this.
--- grischka
Gabor Gyorgy Nagy írta:
Hi list,
seek.c:
#include
#include
#define O_RFLAG GENERIC_READ, FILE_SHARE_READ, NULL, OPEN_EXISTING,
FILE_ATTRIBUTE_NORMAL
, somehow. Looking
into other preprocessor implementations (gcc, pcc, lcc, ...) might
give some ideas.
While trying to feed tcc with output from the GNU preprocessor I fixed this
trivial bug:
Please push patches on our "mob" branch ;)
http://repo.or.cz/w/tinycc.git
Thanks,
--
the development sources at
http://repo.or.cz/w/tinycc.git
Merci,
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
) but tcc
preprocessor expands to CONCAT(X, Y, Z).
The expansion is performed properly when commenting out some nested macro
check code in the preprocessor.
Yes, there are bugs. Here is another one:
#define v s.v
#define S(x) x
S(v)
Should expand to s.v but does to s.s.v
--- grischka
Nikos Mavrogiannopoulos wrote:
grischka wrote:
Nikos Mavrogiannopoulos wrote:
Hello,
I've tried to install tcc on centos 5 (probably old 64 bit version) and
uses some funny paths such as usr/lib64 causing tcc not to actually find
the required files. The attached patch fixes the issue f
n our "mob" branch at http://repo.or.cz/w/tinycc.git
Thanks,
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
ssell.
No matter.
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
:1336: error: ';' expected
There are also some warnings about macro redefinitions but didn't break
the build.
Thanks. I fixed it on the mob branch.
--- grischka
My platform right now is a 32-bit Win7 Professional, with w32api package
version 3.14. TCC was built using a deve
seem to have a different notion of this (that
is the same as tcc, currently) but I think we are safe to assume that
gcc is right ;)
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
it.ufield + sh1) < 0);
> bit.field = -11;
> assert((bit.ufield + bit.field) < 0);
> ch1 = -11;
> assert((bit.ufield + ch1) < 0);
> return 0;
> }
For those who like to play with it, I attached the test cases in a more
runnable form. (run "tcc -D T1..17 -run
:1983).
Is this desired behavior?
Can you show that memory waste actually happens in practice? Say if
you'd load the same .so 1000 times?
--- grischka
Thanks,
Joey
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongn
-run and large distance calls on win64 (as
pointed out recently by Michael Kuklinski)
* Usage of r8/r9 as load/store registers on x86_64
* ...
See http://repo.or.cz/w/tinycc.git
Happy new year,
--- grischka
___
Tinycc-devel mailing list
Tinycc-
is another idea. You could actually just declare the
function(s) as dllexport and then add the program itself using
tcc_add_file. It would then use it's own IAT (import address
table) as jump table.
--- grischka
#include
#include "libtcc.h"
__declspec(dllexport) void print (
ile libtcc with
PE support. Otherwise the generated code would use incompatible
calling convention to functions added using tcc_add_symbol.
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
Michael Kuklinski wrote:
I'm using the master version.
AFAIK there aren't any "cast to uint_32 from void*" in the master
branch.
--- grischka
-Original Message-
From: grischka
Sent: Monday, December 21, 2009 11:33 AM
To: tinycc-devel@nongnu.org
Subject: Re: [
headers (such as stat32/64).
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
possible that
TinyCC is simply compiling the function call with a 32-bit instructon?
master version is here:
http://repo.or.cz/w/tinycc.git
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman
.
That's from 0.9.25? Should be fixed mostly in current master branch.
http://repo.or.cz/w/tinycc.git
Commit patches (if any) directly on our "mob" branch. Thanks,
--- grischka
Dwight
___
Tinycc-devel mailing list
Tinycc-devel@non
http://repo.or.cz/w/tinycc.git/commitdiff/a98caba0
--- grischka
Could you please review the patch?
} else {
if (op == TOK_UDIV || op == TOK_UMOD) {
#if 1
if ((vtop->type.t & VT_BTYPE) & VT_LLONG) {
o(
indow procedures, threads and some such.
--- grischka
#include
#include
#include
int
main(int argc, char *argv[])
{
jmp_buf buf;
if (setjmp(buf) == 0) {
printf("ready to jump at %p\n", &buf);
longjmp(buf, 1);
pr
below).
--- grischka
$ make -C tests btest
btest
--- boundtest 1 ---
succeded as expected
--- boundtest 4 ---
succeded as expected
--- boundtest 8 ---
succeded as expected
--- boundtest 10 ---
succeded as expected
--- boundtest 14 ---
succeded as expected
--- boundt
code representation that tcc does not have.
It's not that I don't like the idea.
---grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
lmost same code. Because nobody wants to fix
the same bugs twice.
Regards, k1w1.
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
Christian Jullien wrote:
Q. how can I see the .s output (as generally using -S on most compilers) ?
TCC outputs machine code, not assembler.
You can disassemble it though:
$ tcc -c x.c
$ objdump -D x.o
on win64:
$ x86_64-pc-mingw32-objdump -D x.o
--- grischka
Christian Jullien wrote:
Well, thanks for explanation; if I was ready to try to fix it, I'm not sure
I have the time to implement it. Sorry.
See http://repo.or.cz/w/tinycc.git/commitdiff/56d6abd
--- grischka
___
Tinycc-devel mailing list
T
some time this year if I found the time and if there
was interest.
Well, becoming a perfect yasm replacement is maybe not the number one
goal for tinycc. ;)
Bill.
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
te that we do not support bounds checking on windows currently.
I don't say it is not possible but it will not work out of the
box.
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
to make a sort
of cross reference of the functions and variables, so you already
(almost) have the prototypes. (See notes in the commit how to use it)
--- grischka
Regards, K1w1.
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://l
ons(TCCState s, const char *option_string);
Which then could be use like:
tcc_set_options(s, "-nostdlib -lsqlite -w -");
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
_PATH=. ./tcc_small -v
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
e bit more what these changes are. ;)
On the other hand, if your plan sounds good, you could just as
well work on the mainline itself.
--- grischka
> Regards, K1w1.
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.or
ent with debug-info IIRC which
might help to find out at least.
--- grischka
Regards.
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
"if (btype.t & VT_TYPEDEF) {")
- Get it back from a typedef symbol into the attribute
(that is in parse_btype() at line "typedef_found = 1;")
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
_CRT_ALIGN(16) for __x86_64 in include/setjmp.h,
but for some reason tcc has problems to remember it.
IIRC it works when used in combination with the variable directly,
such as:
_CRT_ALIGN(16) jmp_buf buf;
or (translating the macro):
__attribute__((aligned(16))) jmp_buf buf;
--- grischka
uot;mob" branch. Also I created a new branch
"mob-stuff" where I stacked some stuff from previous mob that I
don't know what exactly to do with it, at the moment.
Regards,
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel
er than on old tcc-0.9.25. Otherwise nobody will have
time to resolve the merge conflicts.
See: http://repo.or.cz/w/tinycc.git
(There are links to download tar balls for each revision)
Even better if you could push your patches on our mob branch directly.
--- grischka
Hope people will find
Thanks Jörg.
A note to everybody: Be aware that if you configure git with
your real email address it will be visible on the web interface.
--- grischka
Jörg Mische wrote:
Hi all,
I found a small bug when storing structures in vstore(): the store is done by a
call to memcpy() and the
e with tcc (on Windows XP)?
Well, download the package, open with say 7-zip, copy the files
from "include" into your "tcc/include/winapi", and hope that it
works.
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
ht
};
Indeed, there is a missing flcose(op).
I have fixed the leak and attached the file to this mail.
Thanks. Committed to
http://repo.or.cz/w/tinycc.git/commitdiff/cbd7f29e
Please consider sending patches or pushing to our "mob" branch
directly, next time ;)
--- grischka
Best regar
ve further additions or refinements to your
work, feel free to push them on our "mob" branch directly yourself
(It is explained briefly at http://repo.or.cz/w/tinycc.git)
Thanks for your work,
--- grischka
___
Tinycc-devel mailing list
Tinycc-d
if that is what you use.
Is there any
solution on how to compile a program which makes use of any of these calls?
Write them. Or get some source code. (I guess there should be some
in the mingw-runtime-xxx-src pack)
--- grischka
Thanks you very much.
___
Hi, when compiling using GCC the function gets exported as it's named (well,
mangled) but then compiling with TCC its name has an underscore before it:
__declspec(dllexport) LRESULT CALLBACK LowLevelHook(int nCode, WPARAM
wParam, LPARAM lParam)
If GCC it's lowlevelh...@12, TCC makes it _lowlevel
=e90c6f21a89257c8b6e5f396444068e685c1a9e3
I know there are more than one person out there who should be able to fix this, I
just hope it happens ;)
--- grischka
- Original Message -
From: "Rutenis Turcinas"
To:
Sent: Wednesday, October 28, 2009 12:03 PM
Subject: Re: [Tinycc-devel
problems to assemble
its own __chkstk function. In particular there are some 66 prefixes that should
not be there. See below.
Mr. Frederic Feret, would you look into it ?
Thanks,
--- grischka
004010A0 66:872C24 xchg [esp],bp
004010A4 55push ebp
004010A5 66
64, and so on. (not a big
deal but would be nice)
Thanks,
--- grischka
Hi grischka,
> I use TCC to compile TCC. I tried to compile it with PE support under
Windows, and I
obtained some error messages, due to a conflict with the IMAGE_*
structures defined in winnt.h.
So, I renamed IMAGE
_64, and so on. (not a big
deal but would be nice)
Thanks,
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
Floris Bos wrote:
> Altough tcc itself compiles, building libtcc1-x64 fails:
>
>
>
> ==
>
>> .\tcc -o tmp.s -E ../lib/alloca86.S
>
>
>
>> yasm -p gnu -f elf64 -o alloca86.o tmp.s
>
> tmp.s:8: error: invalid size for operand 1
Something got out of sync. It needs to be libtcc1.a
(without "-x64"),
r
- to use a 32 bit compiler to build tcc.exe as cross compiler
(but then you can't use the -run feature).
--- grischka
___
Tinycc-devel mailing list
Tinycc-devel@nongnu.org
http://lists.nongnu.org/mailman/listinfo/tinycc-devel
701 - 800 of 991 matches
Mail list logo
